322 Annals of the Carnegie Museum. 



I have not seen any specimens of Phreatobius cisternarum (Goeldi), 

 but from the description and the habitat assigned I am inclined to 

 believe that it is a very small Pygidium amazoniciim (Steindachner). 



The Breeding Habits of the Geophagin^. 



The Gcophagince have a lobe on the upper branch of the first gill- 

 arch, which is supposed to be associated with the retention of the eggs 

 during the period of incubation. The lobe exists, but it is not used to 

 secure the eggs. I found the eggs of Geophagus hrasilieyisis in the 

 Rio Itapicuru. They were placed in nests similar to those of the 

 North American sun-fishes. The eggs of all the species of this genus 

 are small and numerous, in great contrast to those of species which 

 carry their eggs in the mouth. I took eggs out of the mouth of 

 Tuchysurus barbits, but these eggs were one-fourth to one-half of an 

 inch in diameter, and were contained loosely in the mouth. Besides, 

 I did not see more than about twelve eggs in any one mouth; however, 

 I was told that as many as two dozen are found in large individuals. 

 Osteoglossum bicirrhosum also retains its eggs loosely in a pouch-like 

 depression formed by the elastic skin which connects the bones of 

 the lower jaw. But its eggs are few (less than two dozen) and about 

 one-fourth of an inch in diameter. 



I found one specimen of Geophagus brasiliensis which had a pecu- 

 liarly tuberculated first gill-arch. It appeared to have eggs imbedded 

 between the gill-rakers, but the egg-like structures were abnormal 

 growths. Many species of Geophagus protect their young in their 

 mouths. In the Rio Jacuhy I saw many young Geophagus gymno- 

 genys swimming about a medium-: ized female along a sand-bar. 

 When I attempted to catch them, they nearly all disappeared in the 

 mouth of the mother. When I retreated I saw the young swimming 

 about the mother again. I could not have been deceived, because 

 there were no stones and grass in which they could hide and they do 

 not bury in hard sand. After observing them for some time, I netted 

 the mother, and found her mouth entirely full of young fishes of the 

 same size as those seen swimming about. She began to eject the 

 young a? soon as she was out of the water. I have found eggs in 

 the mouths of various families of fishes, but these eggs either came 

 from ripe ovaries placed in the killing-bucket or were being eaten 

 •or were sucked in during the respiratory process a short time before 



